Meal Map Pro

Meal Map Pro

Puneet Kaur, Harshit, Ritik Saini, Yuvraj Chaudhary, Lokendra Kumar, Komal

Computer science and engineering

Chandigarh University, Mohali,Punjab, India

Abstract-- In this paper, we describe the recipe recommendation system in the culinary domain. Due to the widespread use of the internet, the whole world is connected, and different users from different countries share millions of recipes online, all over the world. As a result, users are unaware of all the recipes available on the internet. A recipe contains heterogeneous information’s such as ingredients, cooking process, categories, etc. Therefore, we believe that a recipe is an aggregation of these heterogeneous features. The majority of the recipe recommendation systems are based on content or collaborate filtering to predict the new recipe that a user is interested in. Combining with both content and collaborative filtering, we propose an effective and elegant framework for combining recipe recommendation systems. Most recipe recommendation systems use content information for ingredients or cooking process of recipes. To reduce RMSE, we proposed a hybrid approach combining conventional techniques with content and collaborative filtering. This approach adds more heterogeneous information such as cuisine, preparation directions, dietary, etc.

Keywords: Recommendation system, collaborative filtering, hybrid approaches, recipes, content information.